Application of double hidden layer process neural network on wood drying modeling

Aiming at time-varying and complex nonlinear characteristics in the wood drying process, a double hidden process neural network was introduced into wood drying modeling. Time-varying and complex nonlinear characteristics of wood drying were considered adequately in wood drying model by time series data of main influence factors fitted to time-varying functions as network input. The concrete modeling process and learning algorithm were given. The effectiveness of method is proved by ash wood drying. It provides a new way for complex system identification.